Abstract
BACKGROUND AND OBJECTIVE: The idiopathic interstitial pneumonias (IIPs) are diffuse parenchymal lung disorders that are associated with substantial morbidity and mortality. Early diagnosis and disease stratification of IIP patients are important because these are related with the treatment and prognosis. Idiopathic pulmonary fibrosis (IPF) and nonspecific interstitial pneumonia (NSIP) are two major distinctive pathologic patterns of pulmonary fibrosis. We researched the application of the fluid-structure interaction (FSI) to the respiratory system and compared the pulmonary acinus mechanics and functions in healthy and IIP models. METHODS: The human pulmonary alveolus is idealized by a three-dimensional honeycomb-like geometry, and a fluid-structure interaction analysis is performed to study the normal and diseased breathing mechanics. The computational domain consists of two generations of alveolar ducts within the pulmonary acinus, with alveolar geometries approximated as closely packed 14-sided polygons. FINDINGS: In a normal breathing cycle, the flow rate of the healthy model is significantly larger than that of the NSIP and IPF models. Similar trends are observed for the volume change and the maximum pressure drop. The flow rate and the volume change of the NSIP are almost the same as those of IPF. The maximum pressure drop of NSIP is 5.5% larger than that of IPF. There is a 47% decrease in the pulmonary acinus compliance for the NSIP and IPF compared with that of the healthy model. The acinus resistances of NSIP and IPF are higher than those of the healthy lung by 6.4~11.2%. In particular, the pulmonary acinus resistance of the NSIP lung is higher than that of the IPF lung by 4.5%. CONCLUSIONS: Our study demonstrates the differences of air flow and lung function in the pulmonary acinus between the healthy and the IIP models. These changes in the lung are important considerations for early diagnosis and disease stratification in patients. Patient-based geometry can to be included in the computational models in future studies.